morign

Morign

AI chat backend with knowledge base Q&A, MCP tools, and OAuth authentication.

Features

• Import documents of knowledge base from a table (CSV), save them into PostgreSQL
• Based on the title and content of the document, generate vector of documents with Embedding API
• Generate document vectors using Embedding API
• Implement high-quality Q&A using vector search
• Welcome message and preset messages
• Chat History for Conversation (based on redis)
• RESTful API
• Support text/event-stream
• Login with OAuth2 client for general Security Provider
• Built-in MCP (Model Context Protocol) tool support

Supported Frontend

APIs

Get session information

Get user information of that has been verified or signed in

Post chat prompt and return Streaming messages

Getting started

# Install dependencies (includes forego for env loading)
make deps

# Or manually
go mod tidy
go install github.com/ddollar/forego@latest

# Create .env from example and configure (update PG/Redis URLs, API keys, etc.)
test -e .env || cp .env.example .env

# Start server with environment variables from .env
forego start

# Or directly (for development)
forego run go run . web

Prepare preset data file in YAML

# Simple preset
welcome: "Hello, I am your virtual assistant. How can I help you?"
systemPrompt: "You are a helpful assistant."
toolsPrompt: "You will select the appropriate tool based on the user's question and call the tool to solve the problem."

# Or with custom tool descriptions
welcome: "Hello, I am your virtual assistant. How can I help you?"
systemPrompt: "You are a helpful assistant."
toolsPrompt: "You will select the appropriate tool based on the user's question and call the tool to solve the problem."
tools:
  kb_search: "Search documents in knowledge base with subject. When faced with unknown or uncertain issues, prioritize consulting the knowledge base."
  kb_create: "Create new document of knowledge base, all parameters are required. Note: Unless the user explicitly requests supplementary content, do not invoke it."
  fetch: "Fetches a URL from the internet and optionally extracts its contents as markdown"
  memory_list: "List all stored memories"
  memory_recall: "Search memories by keyword"
  memory_store: "Store a new memory"
  memory_forget: "Delete a memory by key"

• welcome: Welcome message displayed to users
• systemPrompt: System prompt for AI conversation
• toolsPrompt: Instructions for tool usage (used when MCP tools are available)
• tools: Custom tool descriptions (optional, overrides built-in defaults) Note: Memory tools (memory_*) are bound to and isolated by the logged-in user identity.

See data/preset.example.yaml for a complete example.

Prepare database

CREATE USER morign WITH LOGIN PASSWORD 'mydbusersecret';
CREATE DATABASE morign WITH OWNER = morign ENCODING = 'UTF8';
GRANT ALL PRIVILEGES ON DATABASE morign to morign;


\c morign

-- optional: install extension from https://github.com/pgvector/pgvector
CREATE EXTENSION vector;

Command line usage

USAGE:
   morign [global options] command [command options] [arguments...]

COMMANDS:
   usage, env                   show usage
   initdb                       init database schema
   import                       import documents from a csv
   export                       export documents to csv/jsonl
   embedding, embedding-prompt  read prompt documents and embedding
   agent, llm, chat            test LLM agent
   web, run                     run a web server
   version, ver                 show version
   help, h                      Shows a list of commands or help for one command

GLOBAL OPTIONS:
   --help, -h  show help

Agent Command

Test LLM functionality from command line:

# Non-streaming chat
./morign agent -m "hello"

# Streaming chat
./morign agent -m "hello" -s

# Show verbose logs
./morign agent -m "hello" -v

Parameters:

• -m, --message: message to send (required)
• -s, --stream: enable streaming response
• -v, --verbose: show logs (disabled by default)

Change settings with environment

Show all local settings

./morign usage

# HTTP server
MORIGN_HTTP_LISTEN=:3002

# optional preset data
MORIGN_PRESET_FILE=./data/preset.yaml

# optional OAuth2 login
MORIGN_AUTH_REQUIRED=true
OAUTH_PREFIX=https://portal.my-company.xyz

# optional proxy
HTTPS_PROXY=socks5://proxy.my-company.xyz:1081

Environment Variables

Provider Configuration (AI Services)

Each provider requires API_KEY and MODEL, optional URL and TYPE for custom endpoints:

Supported Provider Type: openai, anthropic, openrouter, ollama

Example:

# Interact provider (supports openai/anthropic/openrouter/ollama)
MORIGN_INTERACT_API_KEY=sk-xxx
MORIGN_INTERACT_MODEL=gpt-4o-mini
MORIGN_INTERACT_TYPE=openai  # optional, default openai

# Using Anthropic
MORIGN_INTERACT_TYPE=anthropic
MORIGN_INTERACT_MODEL=claude-3-5-sonnet

# Embedding provider
MORIGN_EMBEDDING_API_KEY=sk-xxx
MORIGN_EMBEDDING_MODEL=text-embedding-3-small

# Summarize provider
MORIGN_SUMMARIZE_API_KEY=sk-xxx
MORIGN_SUMMARIZE_MODEL=gpt-4o-mini

Tip: Run ./morign usage to view all current configurations

The operation steps for generating data.

1. Prepare a CSV file for the corpus document.
2. Import documents.
3. Generate Questions and Answers from documents with Completion.
4. Generate Prompts and vector from QAs with Embedding
5. Done and go to chat

CSV template of documents

./morign initdb
./morign import mycompany.csv
./morign embedding

Attach frontend resources

1. Go to frontend project directory
2. Build frontend pages and accompanying static resources.
3. Copy them into ./htdocs

Example:

cd ../Calisyn
npm run build
rsync -a --delete dist/* ../morign/htdocs/
cd -

During the development and debugging phase, you can still use with proxy to collaborate with the front-end project.